Content Measurement Template

Working out what your content actually did, when attribution is genuinely hard.

Start with the honest part

You cannot attribute most content precisely, and pretending otherwise produces confident wrong decisions.

Someone reads three of your posts over two months, hears you mentioned once, then searches your name and buys. Your analytics will record "direct traffic" and credit nothing. This is the normal case, not the edge case.

So measure two things separately: what you can attribute, and what you can only observe.

What you can attribute

Signal	How	Reliability
Link clicks with a tracking parameter	UTM tags	High for the click, low for the journey
Signups from a dedicated landing page	One page per campaign	High
Replies and DMs referencing a piece	Manual log	High, and undercounted
"How did you hear about us?"	Ask at signup	Medium — people misremember, but directionally useful
Discount or referral codes	Unique per channel	High

What you can only observe

Signal	What it suggests
Branded search volume rising	People are looking for you by name
Direct traffic rising after publishing	Attribution you cannot see
Inbound enquiries mentioning your topics	Your positioning is landing
People quoting you back to you	Real reach
Sales cycles shortening	Prospects arrive pre-convinced

That last one is the most valuable and least measurable effect of good content. Track it anyway.

The tracking sheet

Date	Piece	Channel	Reach	Clicks	Meaningful responses	Led anywhere	Time spent

Meaningful responses — a reply with substance, a DM, someone quoting it — is the column that predicts business outcomes. Reach does not.

Time spent is the column everyone omits and the one that tells you whether a format is worth continuing.

UTM discipline

If you use tracking parameters, be consistent or the data is worse than none.

utm_source   = where it was published    (linkedin, newsletter, youtube)
utm_medium   = the type                  (social, email, video)
utm_campaign = the piece                 (2026-09-pricing-post)

Lowercase, no spaces, same spelling every time. LinkedIn and linkedin become two rows in every report you ever run.

The monthly analysis

Here is my content performance for [MONTH]:
[PASTE THE TABLE]

1. Which pieces produced meaningful responses, not just reach?
2. Is there a pattern in subject, format, angle or length?
3. Which took the most time relative to what they returned?
4. Where reach was high but responses were zero, what does that suggest?
5. What should I stop doing?

Do not tell me to post more consistently or to keep experimenting.
Be specific about what the data supports.

Question 4 matters: high reach with no response usually means the piece was agreeable rather than useful. Agreement does not produce conversation.

Prediction, used carefully

Here are my last 20 pieces with their performance: [PASTE]

What features correlate with the pieces that produced meaningful responses?
Consider: topic, format, length, whether it took a position, whether it
contained a specific number, whether it described a failure.

Be honest about sample size. With 20 pieces, tell me which patterns are
plausible and which could easily be noise.

The honesty instruction matters. With twenty data points, almost nothing is statistically meaningful, and AI will happily produce confident patterns anyway.

The traps

Optimising for the measurable. Reach is easy to measure, so people optimise for it and end up writing agreeable, forgettable content that reaches many people and moves none.

Over-attributing. One post preceded a good month, so the post caused the good month. Usually it did not.

Ignoring time cost. A format with twice the engagement and four times the effort is a worse format.

Too-short horizons. Content compounds. Judging a piece after 48 hours misses most of what it does — especially anything searchable.

Chasing an outlier. One piece did unusually well. Trying to reproduce it directly usually fails; the useful question is what was true about it, not what it said.

The quarterly question

Once a quarter, ask the only question that really matters:

Is this content producing the business outcome I wanted, or am I producing content because I have a system for producing content?

If the honest answer is the second, change the channel, the audience or the format. Do not just post more.

Quarter	Business outcome sought	Evidence it is happening	Decision